Program > Papers by speaker > Dragun Kirill

The prices of the traded assets that are comprised in an ETF do not update simultaneously and this asynchronicity creates difficulties when estimating covariances among assets. We propose the Beta Adjusted Covariance estimator as an improvement to a traditional realized covariance estimator by exploiting the information in the realized stock-ETF beta (i.e., the covariance between the ETF's components and the ETF itself) estimated at the highest frequency possible. We find that the proposed estimator efficiently deals with biased approximations by traditional estimators caused by asynchronous trading data and significantly improves accuracy of the estimated covariances.